1. Field of the Invention
The present invention relates to a technique of converting tone image data into a form expressed by the dot on-off state, and more particularly to a technique for the rapid conversion of image data with sustained quality.
2. Description of the Related Art
Image display apparatus in which images can be created by the formation of dots on display media such as print media or liquid-crystal screens are widely used as the output devices of various types of image-output equipment. The operation of such image display apparatus is limited to the creation of images by dots formed locally, but images with continuously varying its tones can be created by appropriately controlling the formation density of dots in accordance with the tone values of the images.
A method referred to as an error diffusion technique is widely used for these image display apparatus in order to determine the dot on-off state at each pixel so that the dots are formed with in a density adequately selected in accordance with the tone values of the image. The error diffusion technique is a method whereby the binarization errors generated by deciding the dot on-off state at a target pixel are diffused among the unconverted pixels on the periphery of the target pixel, the result is stored, and the dot on-off state is determined such that the errors diffused from the peripheral pixels are eliminated when the dot on-off state is determined for the unconverted pixels. Thus, the formation of dots whose density is sufficient to conform to the tone value of an image can be ascertained by determining the dot on-off state in a manner that allows the binarization errors generated by the peripheral pixels to be eliminated.
Using the error diffusion technique makes it possible to create high-quality images because dots are formed at an optimum density suitable for the tone image data, but because the tone errors must be diffused among peripheral pixels every time the dot on-off state is determined, processing time increases and rapid image expression becomes difficult to accomplish with an increase in the number of pixels in the images. To address these problems, techniques have been proposed in which a specific number of adjacent pixels is combined into a block, and the dot on-off state is determined while the errors are diffused from this block to adjacent blocks (see, for example, JP2000-22944A). Determining the dot on-off state in block units in such a manner allows processing to be completed in a short time and images to be rapidly displayed even when a larger number of pixels is to be processed.
Such methods are disadvantageous, however, in the sense that the quality of displayed images tends to be adversely affected when these methods are used. This is attributed to the fact that determining the dot on-off state in block units obtained by combining a specific number of pixels amounts to reducing image resolution and tends to reduce image quality in proportion to the reduction in resolution.
An purpose of the present invention, which was perfected in order to address the above-described problems of the prior art, is to provide a technique that allows images to be rapidly converted to an expression format based on the dot on-off state while preserving image quality.